Please feel free to jump into the Methods and Analysis sections if you are not a newbie in crypto… for those of you who are I recommend enjoying the full article…
Introduction
At present the entire cryptocurrency market is driven by bitcoin (BTC) and its halving events which leads to approximately 4-year market cycles. Bitcoin halving is when the rate of supply of new BTC is cut in half which happens approximately every 4 years (210,000 blocks). As a result, the asset becomes scarcer and when you analyse the price history of BTC you can see the BTC price rallying after these halving events and leading towards a new market cycle.
So, why does the entire crypto market follow BTC?
Being new to crypto this was a question which was troubling me…. wont this be a disadvantage for the other crypto projects?… they won’t be judged / evaluated solely based on their utility. I decided to explore some on-chain metrics and the answer became clearer. BTC has a large network effect (e.g., BTC have 16.5 million active addresses compared to 6.5 million in Ethereum (ETH) — September 2021 [Glassnode]. Active addresses — the number of unique wallet addresses active in the network either as the sender or receiver. My initial guess was ETH would have more active wallets due to its ecosystem). So, to conclude, BTC helps other altcoins gain attention and exposure and helps them grow. However, in the long run when the adoption dilutes among different crypto currencies it could possibly de-correlate the crypto market up to an extend and reduce the volatility of crypto portfolios.
Market Cycles
When an BTC halving occurs the new supply of BTC drops which helps increase the value of the asset. If you have not already done so, I recommend you read the articles by PlanB on Medium which analyses this and presents the Stock to Flow (S2F) and Stock to Flow Cross Asset (S2FX) models, which will help you get a clear understanding [1, 2, 3]. There have been three BTC having events so far (Nov 2012, Jul 2016, May 2020). During the first two cycles the price of BTC increased by 93x and 28x respectively, and soon after the prices fell by 84% and 83% respectively [4]. Currently we are in the middle of the third cycle.
The patterns we see in financial markets has a lot to do with the psychology of the market participants. It is very interesting how we can analyse these psychological aspects through technical analysis tools [5]. My focus was on trying to use the on-chain data to analyse these psychological aspects.
Specifically, I was interested in trying to understand how and when the markets would react to these halving events.
Methods
I used BTC data from coinmetrics [6] and my code can be found on GitHub [7]. My focus was on Price and Active Address data. A summary of the data is presented in Figure 1, along with the bitcoin halving events highlighted.
Results & Analysis
From Figure 1, we can see that the increase in active addresses (network activity) leads towards price peaks of the cycles. Which is the expected behaviour. And soon after the peak of the bull market, the price drops along with the active wallet addresses. In my opinion during a bull run a portion of the new users entering crypto… are only there due to the speculation and quick profits… not necessarily attracted to the utility… and these people eventually exits the market… however, on a positive note after each cycle the user base who believes in BTC increase.
Even though the increase in active addresses during a bull run is visual and intuitive in Figure 1, the slow increase in active addresses in anticipation of an upcoming halving event was not. I was more interested to see during which periods the market would start to prepare for the next bull run (or the halving event). Thus, I formulated the following metric, which simply focuses on the percentage change in the active addresses and normalises it by the percentage price change I use the 360-day Simple Moving Average (SMA) of the two target variables. The result of the proposed metric is shown in Figure 2. (I have not conducted a literature review so I’m not sure if this metric exists… this was just a Sunday hobby project).
For simplicity let’s call this metric “AACPC Ratio” — Active Address Change per Price Change Ratio (Not the best name… but it will do for now ).
The results were very interesting, where you can observe two main spike regions in the AACPC Ratio ahead of all the three halving events. Let’s define the following abbreviations for clarity of the analysis,
S1 — Spike closest to the halving event
S1HD — Time period between S1 and future halving event
S2 — The second main spike
S2HD — Time period between S2 and future halving event
First let’s analyse S1. S1 could represent the market participants preparation for the next bull market, and it is interesting to see that S1HD periods are lengthening or i.e., the participants are getting ready in advance (1st Cycle — 3 months, 2nd — 7 months, 3rd — 11 months). We can speculate that the lengthening of the period could be due to more informed market participants. The lengthening of S1HD could also be due to diminishing returns and could eventually lead to lengthening market cycles.
The reason behind S2 isn’t straightforward and seems to represent the current market cycle bottom. Further exploration is required in this regard and please let me know your thoughts in the comments section.
The AACPC Ratio of the current cycle is very interesting as we can see two spikes at the very beginning of the current market cycle which was not present in the previous cycles. This indicates that the change in active addresses is greater than the price change during the early stages of this cycle. What does this break in the pattern tell us? I look forward to analyse this further in my next article. Most probably it could be that a lot of people FOMO’ed into the market right after the halving, which could be a possibility as the amplitude of the spikes prior to the 2021 bull run seems to be smaller than the previous two cycles (Further analysis is required considering the magnitude of active address and price).
Conclusion
The use of the AACPC Ratio paves way for an interesting analysis. However, we are limited to data of only two prior market cycles. Monitoring this metric would help us understand the sentiment in the market and understand the potential towards an upcoming bull run and time the entry.
I would love to hear your thoughts and discuss about this idea to explore the merits and flaws..
If you enjoyed this article and like to read more, please follow me on medium and twitter and share this article.
References:
1. Modeling Bitcoin Value with Scarcity — https://medium.com/@100trillionUSD/modeling-bitcoins-value-with-scarcity-91fa0fc03e25
2. Efficient Market Hypothesis and Bitcoin Stock-to-Flow Model- https://medium.com/@100trillionUSD/efficient-market-hypothesis-and-bitcoin-stock-to-flow-model-db17f40e6107
3. Bitcoin Stock-to-Flow Cross Asset Model -https://medium.com/@100trillionUSD/bitcoin-stock-to-flow-cross-asset-model-50d260feed12
4. Bitcoin Investor Guide — Standard Chartered — https://www.tbstat.com/wp/uploads/2021/09/Bitcoin-investor-guide.pdf
5. Technical Analysis of the Financial Markets (John J. Murphy)
6. Data source: https://coinmetrics.io/community-network-data/
7. Code: https://github.com/s1mplev99/BTC_MarketCycle_Analysis
